The number of adolescents and adults with eating disorders has been increasing, and it has been reported that up to one third of elementary school children are preoccupied with dieting and weight. Although longitudinal research suggests that feeding problems and eating disorders often start early and are stable over time, little is known about how the regulation of eating develops in infants and young children. It is proposed that the regulation of eating in infancy develops in the context of parent-infant interactions. Feeding disorders develop when specific infant and parent vulnerabilities interact and difficulties arise during feeding. The specific aims of this project are twofold: 1) To further validate this transactional model for the regulation of eating in infants and young children; 2) To test the effectiveness of a short-term intervention based on this transactional model for a specific feeding disorder-Infantile Anorexia. Infantile Anorexia is characterized by the onset of persistent food refusal during the transition to self-feeding, malnutrition, parental anxiety about the infant's poor food intake, and mother-infant conflict during feeding. It is the most severe form of a spectrum of feeding problems in infants and young children. A partial dismantling design has been chosen to test the transactional model and the efficacy of a brief intervention, based on this model, to treat infantile anorexia. There will be two treatment groups with 55 infants in each group. The infants will be randomly assigned to one of two interventions. Intervention Group I will receive the full treatment model: a Psycho-Educational Treatment which will address the infant's temperament and the parent's vulnerability in regards to limit setting. This intervention will prepare the parents for the second part of the intervention, the Parent Training, which gives the parents specific suggestions on feeding routines and behavior management of the infant in order to facilitate internal regulation of eating. Intervention Group II will receive Parent Training plus four hours of therapist contact which serves as a Control Condition. This Control Condition was chosen to be as non-specific and as different as possible from the Psycho-Educational Treatment. If, as hypothesized, the Psycho-Educational Treatment plus Parent Training brings the greatest improvement in parent-infant interactions and thus facilitates the infant's eating and gaining weight, these data will further validate the transactional model for Infantile Anorexia. In addition, these data will clarify how to facilitate internal regulation of eating in infants and young children in general. The principles of internal regulation of eating can be applied to the prevention and treatment of other eating disorders of children and adolescents as well.